Impacts of brush pile fishing on fish biodiversity: A case study of the Shari-Goyain River in Bangladesh

Abstract

Brush pile (katha), a fish aggregating device, has been widely used in the Shari-Goyain River since 2003 to congregate fish for easier catch. Katha is usually used during the winter season when the water depth decreases. Hence, this experiment was conducted from November 2018 to March 2019 on katha fishing to investigate its status and impacts on fisheries resources of the Shari-Goyain River in the Sylhet district of Bangladesh. The study was based on the hypothesis that katha fishing might have detrimental impacts to fish biodiversity and production. Data were obtained through a questionnaire-based survey, personal interviews, catch assessment (CA), focus group discussions, and key informant interviews. A total of 54 species were documented, including two exotic fish species (tilapia and common carp) and 3 species of prawn during harvesting of the kathas. The catch per unit effort (CPUE) (kg/gear/ha/person/hour) was the highest in December (1.13 ± 0.37), followed by November (1.06 ± 0.40), January (0.80 ± 0.25), February (0.71 ± 0.23), and March (0.52 ± 0.21). The catch per unit area (CPUA) (kg/ha) was the highest in November (264.66 ± 18.21), followed by December (205.05 ± 27.77), January (175.02 ± 76.04), February (147.73 ± 52.11), and March (102.08 ± 41.04) where significant differences (p < 0.05) among the months were observed. Average catch per katha in a month ranged from 41.09 ± 16.11 to 12.42 ± 5.89 kg, with a mean of 24.29 ± 11.08 kg, and a significant decrease in average catch was observed with the progression of months. The most species richness was noticed in December (38), followed by November (35), January (34), February (28), and March (25). Siluriformes (39.123%) was the most dominant order, followed by Cypriniformes (33.956%), Decapoda (14.661%), and Ovalentaria (3.278%). According to the CA and respondents’ perception, indiscriminate harvesting of fish by katha fishing can be a cause of fish biodiversity loss as it reduces open water catches, total production, and disturbs the ecosystem. From the research findings, it is suggested that katha fishing should be stopped for sustainable management and conservation of fisheries resources in the Shari-Goyain River. Research on the effects of katha fishing should be conducted in other open waters of Bangladesh where this type of fishing is common.

Katha fishing, Fish diversity, Production, Riverine fishes, Management

1. Introduction

Bangladesh is an important inland fishery resourceful country, blessed with a large number of inland waterbodies like rivers, freshwater marshes, canals, brackish water impoundments, natural and manmade lakes, beels, haors, and floodplains (DoF, 2020; Saha et al., 2021). It has the third largest aquatic biodiversity in Asia, and the presence of the world’s largest flooded wetland makes the country one of the most suitable areas for fish in the world (Shamsuzzaman et al., 2017). However, illegal fishing and fishing pressure on the aquatic ecosystem are increasing due to the rapid increase in the human population and the growing demand for animal protein. Hence, the annual harvests from the rivers and estuaries are on increasing trends (DoF, 2009; DoF, 2020). In 2007–08, the annual fish production from rivers and estuaries was only 1,36,812 MT (DoF, 2009). The riverine production of fish increased from 3,25,478 MT in 2018–19 to 3,31,793 MT in 2019–20 with a growth rate of 1.94% (DoF, 2020). As a result, overharvesting of fish using illegal fishing gears and indiscriminate methods of fishing is very common in waterbodies, particularly in the inland open waters of Bangladesh (Galib et al., 2009; Sufian et al., 2017; Arefin et al., 2018; Akter et al., 2020; Pandit et al., 2020, 2021, 2022; Saha et al., 2021; Mia et al., 2022). Consequently, 64 indigenous fish species of Bangladesh have become red listed as threatened, where 52% of those are riverine species (IUCN Bangladesh, 2015). Thus, the biodiversity of riverine fishes is in great danger and many of them are vulnerable, endangered, or critically endangered. Therefore, it is necessary to find out the destructive fishing methods responsible for overfishing and indiscriminate killing of fish.

Katha or brush pile fishing is a traditional fishing method widely used in the rivers and other natural wetlands of Bangladesh. In general, traditional katha fishing is not a destructive method of fishing. However, nowadays, various destructive gears and methods are used during the harvesting of katha. As a result, it needs to be explored whether katha fishing is truly harmful to fish biodiversity. The term katha has many Bengali synonyms, such as katta, jhag, jhata, etc. It is also called komar when operated in oxbow lakes (Middendorp et al., 1996). Katha is also known as Fish Aggregating Device (FAD), brush pile, brush shelter, or brush park (Mustafa, 2017; Uddin et al., 2015). Katha acts like a short-time shelter for fish and works as a nursing and feeding ground for them. Moreover, fish also use katha as a hiding place and protection from predators. Therefore, schools of various fish species accumulate in katha, which makes it a fish aggregating device in freshwater environments, similar to the marine FAD (Cressey, 2014). In freshwater katha preparation, substrates like branches of bamboo, hijol(Barringtonia acutangula), koroch (Millettia pinnata), mango (Mangifera indica), blackberry (Syzygium cumini), etc. are used as a medium for shelter and algal attachment. Thus, katha is a manmade artificial object or brush park suspended in the water column and fixed to the bottom to attract fishes so that they are aggregated for the purpose of shelter, food, protection from predators, nursing, breeding, and other purposes (van Dam et al., 2002; Wahab and Kibria, 1994). Different species of fish at different ages take shelter inside katha. Its structure materials are usually selected according to the preferences of species by analyzing their behavior and characteristics. Usually, katha is established and operated during the dry season (November to March). Joadder et al. (2016) mentioned katha fishing as prohibited method of fishing according to the fisheries regulation which were practiced in the Beel Kumari of Bangladesh. Katha usually ranges from 6-9 m in length, 2–6 m in width, and is installed along the edge of waterbodies with a depth of around 1.25 m (van Dam et al., 2002). During harvesting, the entire katha is enclosed with a fine meshed net to make sure that no fish can escape, and then the brush piles are taken out from the net enclosure (van Dam et al., 2002). Afterwards, harvesting is done with cast net, wounding gears, hand picking, push net, and finally with the net that encloses the katha. Small non-mechanized boats called dinghi are used inside the katha for fishing. The size of katha usually ranges from 0.12 to 1.17 ha, with an average of 0.35 ha (Ahmed and Akther, 2008). Each of the katha is fished for around 2–3 times a year, especially in the lean season (Mustafa, 2017). Fishing pressure has been increased in the rivers due to the establishment of katha and its indiscriminate harvesting. Both fishers and non-fishers place katha in the river. Unplanned and unregulated use of this fishing device causes a serious threat both to the natural stocks and to the effectiveness of stock enhancement as all kinds of fish of different sizes and ages are harvested (Galib et al., 2009). Thus, katha fishing in the river or any waterbody has a detrimental impact on fisheries resources, also reducing fishing opportunities for poor and marginal fishers (Mustafa, 2017).

The Shari-Goyain River is a transboundary river that originates from the Meghalaya hills of India and enters Bangladesh through the northern part of Jaintiapur upazila of Sylhet district and meets with the Surma River near Chhatak upazila in Sunamganj district (Talukder et al., 2021). The average width of this river is about 100 m (Shumi et al., 2019). Thousands of families are directly involved in fishing in this river for their livelihood (Shumi et al., 2019). Various types of fishing gears and methods are used for the harvesting of fish by the fishers. However, according to the fishers’ statements, katha fishing has been widely used in the Shari-Goyain River since 2003. At present, katha fishing has become very common in the Shari-Goyain River during the dry season. In the dry season, water flow is reduced to a precise level, and the river serves as a reserve for a variety of aquatic species that are important for breeding in next spawning seasons. Hundreds of kathas are constructed in the river during this time, and fishers harvest fishes as much as they can, including juveniles and brood fishes, which accumulate in the kathas as their preferred shelter. This sort of indiscriminate fishing has become more popular in the river, which could have a serious impact on natural fish supplies from the river. However, determination of the damage caused by katha fishing and saving the riverine environment are extremely challenging. Thus, the present study was conducted in the Shari-Goyain River to assess the present status of katha fishing and its impact on the fisheries resources of the river. This research has potential to assist wetland management authorities in taking steps to regulate katha fishing in order to protect the riverine fisheries resources of Bangladesh.

2. Materials and methods

2.1. Selection of study period and area

In the Shari-Goyain River, kathas are usually established at the end of the rainy season when the water level starts getting lower and are usually removed at the beginning of the rainy season. Therefore, this study was conducted from November 2018 to March 2019 as katha fishing was performed during this period. Each katha was operated for an average of 2.21 times per year, shifting the place of establishment to a nearby area of the river. With a frequent visit to the whole river the study area was selected from Gowainghat to Salutikor Bazar, covering an area of 27 km along the length of the Shari-Goyain River (Figure 1).

Figure 1.

Map of the study area showing in light-green colour (Source: Google Earth Pro).

2.2. Description of the katha fishing method

Katha is a manmade artificial structure (Figure 2) where different species and ages of fish take shelter. During harvesting, the entire katha is surrounded by a fine-mesh synthetic net to ensure that fishes cannot escape and the brush piles are subsequently removed from the net enclosure. Inside the katha small non-mechanized boats called dinghi are used for fishing. Cast nets, wounding gears, manual picking, push nets, and lastly the net that encloses the katha are used to harvest entire fishes.

Figure 2.

A schematic diagram of traditional katha in the Shari-Goyain River.

2.3. Selection of target group

A large number of fishermen and people from different walks of life are engaged in katha fishing activities. Many non-fishers also establish katha and invest money in the preparation of katha. These people are known as katha owners. A total of 56 fisher and non-fisher katha owners were interviewed from Jalurmukh, Motorghat, Dariakandi, Salutikor, Alinagar, Gowainghat, Meuarkandi, Satkulikandi, and Kachuarpar villages of the Nandirgaon union and Purnanagar, Mugambari, Lengura, and Nihain villages of the Alirgaon union. These villages are situated on the banks of the Shari-Goyain River. The fishermen are highly dependent on the river for their livelihood.

2.4. Preparation of the questionnaire

A draft questionnaire was developed for pre-testing by a few of the respondents. Based on the results gathered from the pre-test, the questionnaire was then modified and rearranged. Katha materials, fishing gears, fishing crafts, fish availability, catch composition, amount and number of fish caught, etc. were included in the questionnaire. During data collection, it was confirmed that informed consent was obtained from all survey participants of the present study.

2.5. Data collection procedure

Primary data were collected from katha fishers, local fish traders, and katha owners through Catch Assessment (CA), Personal Interview (PI), Focus Group Discussion (FGD), and Key Informant Interview (KII). Upazila Fisheries Officers (UFOs), experienced people, and local leaders were interviewed as key informants. In FGDs, fishers, katha owners, fish traders, and local people were encouraged to gather in a place and talk about river fishery, katha fishing, biodiversity of fishes, the causes of fish depletion, etc. Each of the FGDs was organized with 8–15 members. A total of 56 fishers and non-fisher katha owners were interviewed during catch data collection and recorded as PI. All of the existing kathas in the Shari-Goyain River were counted by a direct visit to the entire study area. Moreover, the number of operations of each katha in the entire study area was also recorded.

2.6. Catch assessment procedure

Catch assessment was performed four times per month during the study period. Fishers were interviewed about their catch, previous catches, available fish species, and abundance of the fish. Normally, katha fishing starts at 7:00 to 10:00 am and finishes at 3:00 to 4:00 pm. Sometimes, it takes more time depending on the size of the katha. A group of fishermen consisting of 8–15 members encircled the entire katha with a fine meshed long seine net and the katha materials such as bamboo, hijol, koroch, etc. were gradually removed. Harvesting of katha was done by hand picking, cast net, push net, drag net and finally using the encircled seine net. Catch per unit area (CPUA) and catch per unit effort (CPUE) were used as a measurement of fish production in the total area of katha (Mustafa, 2017).

$$\text{CPUA}=\frac{\text{Total catch }\left(\text{kg}\right)}{\text{Total area of }katha\left(\text{ha}\right)}$$ (1)

$$\text{CPUE}=\left\{\text{Total catch }\left(\text{kg}\right)\right\}/\left\{\right(\text{Number of gears }\times \text{total area of }katha\left(\text{ha}\right)\times \text{number of fishermen }\times \text{total fishing time }\left(\text{hour}\right)\}$$ (2)

The name of the fish species, the number and weight of fishes were recorded on the spot during the katha fishing. In the case of small fishes, a portion of catches comprising small fishes were weighed and counted. Thus, the total weight and total number of fishes were estimated. For medium-sized fish, a total of 4–5 fish were taken for each species. The average katha catch of each month was recorded and compared. By compiling the monthly harvest of all the katha, the total catch of katha for that season was estimated.

2.7. Total catch (kg) estimation of the katha

$$\text{Total catch of a }katha\left(\text{kg}\right)=\text{Average }katha\text{ catch }\left(\text{kg}\right)\times \text{total number of harvests in a season}$$ (3)

$$\text{Total production }\left(\text{kg}\right)=\text{Total catch of a }katha\left(\text{kg}\right)\times \text{total number of }katha$$ (4)

$$\text{Average }katha\text{ catch per month }\left(\text{kg}\right)=\text{Total catch in a month }\left(\text{kg}\right)\text{ }/\mathrm{t}\text{otal number of fishing days}\text{ in a month}$$ (5)

2.8. Fish species diversity indices

Biodiversity indices such as the Shannon-Weiner diversity index (H′) (Shannon and Weiner, 1949), Margalef’s richness index (d) (Margalef, 1968), Simpson’s dominance index (D) (Simpson, 1949), and Pielou’s evenness index (J) (Pielou, 1966) were calculated.

The Shannon-Weiner index (H′) was calculated as:

$$\mathit{\text{H}}\text{'}=-\sum _{i=1}^s{p}_i\ln p_i$$ (6)

Where,H' = Shannon-Weiner diversity index,

S = Number of species,

p_i = n_i/N,

n_i = Number of individuals of a species, and.

N = Total number of individuals.

Margalef’s richness index (d) was calculated by this formula:

$$\mathit{\text{d}}=(S-1)/\log \left(N\right)$$ (7)

Where,

S = Total number of species and

N = Total number of individuals.

Simpson’s dominance index was calculated by the following formula:

$$\mathit{\text{D}}=1-\left(\frac{\sum n(n-1)}{N(N-1)}\right)$$ (8)

Where

n = The total number of individuals of a species and

N = The total individuals of all species.

Pielou’s evenness index (J) is defined as:

J = H_(S)/H_(max) (9)

Where,

H_(S) = The Shannon-Weiner diversity index and

H_(max) = The maximum possible value of the Shannon-Weiner index if all the values are identical.

2.9. Statistical analyses

Collected data were edited, coded, and computed for analysis. These data were verified to eliminate all possible errors. For instance the normality of the data, their homogeneity, and independence of units were checked. For the processing of data, a tabular technique was applied by using simple statistical tools like averages and percentages. Data were analyzed using Microsoft Office Excel 2010, as well as the Statistical Package for the Social Sciences (SPSS version 20). One-way Analysis of Variance (ANOVA) was used for statistical analysis at 5% level of significance. Finally, Kruskal-Wallis ANOVA by ranks (Fakayode et al., 2012) was used to evaluate the negative impact of katha fishing.

3. Results

3.1. Status of katha and katha fishing in the Shari-Goyain River

A total of 187 kathas were found in the study area, and each of the kathas was harvested 2.21 ± 0.78 times a year. It was estimated that total kathas were harvested 413 times during the study period, of which 76 catches were sampled for the current study. The average katha size in the study site was 0.13 ± 0.06 ha. The mean construction cost of an individual katha was 7,857.14 ± 5,536.57 Bangladeshi Taka (BDT), depending on its size. The benefit-cost ratio of a single katha was 1.11. The average group size of the katha fishers was 10.2 ± 2.9 person. The average fishing time of kathas was 5.97 ± 1.2 h. Dinghi, with an average length of 8.28 ± 1.21 m and a width of 1.40 ± 0.49 m, were used to operate the net and other fishing gears. The boats are also used to keep the harvested fishes. Notably, most of the katha owners were to be found non-fishers, and they got around 50% share of the total katha catch. A fisherman can earn approximately 15,000 ± 350 BDT per month during the katha fishing season.

3.2. The diversity and abundance of aquatic species in the katha

A total of 54 species were recorded, including two exotic species (Oreochromis niloticus, Cyprinus carpio) and 3 species of prawn under 3 classes, 14 orders, and 23 families from the katha during the harvesting period (Table 1). Siluriformes was the most dominant order, constituting about 39.123% of the total catch, followed by Cypriniformes (33.956%), Decapoda (14.661%), Ovalentaria (3.278%), Tetraodontiformes (3.186%), Synbranchiformes (2.068%), Beloniformes (1.032%), Clupeiformes (0.857%), and the rest 1.837% were comprised by Gobiiformes, Cichliformes, Mugiliformes, Osteoglossiformes, Anabantiformes, and Cyprinidontiformes. Among all the species, Macrobrachium malcolmsonii was the most abundant species in the capture (relative abundance 37.179%), followed by M. lamarrei (relative abundance 30.027%) and Parambassis ranga (relative abundance 3.427%). On the other hand, as a single species, Wallago attu dominated the catch by weight and contributed 13.098% to the total catch.

Table 1.

Species composition of fishes from katha fishing in the Shari-Goyain River.

Order	Family	Local name	Scientific name	IUCN status	Relative abundance (%)	Catch composition (%)
Siluriformes	Bagridae	Golsha	Mystus cavasius (Hamilton, 1822)	NT	1.968	7.115
		Rita	Rita (Hamilton, 1822)	EN	0.062	1.365
		Ghagla	Hemibagrus menoda (Hamilton, 1822)	NT	0.063	3.809
		Air	Sperata seenghala (Shykes, 1839)	VU	0.073	6.989
		Guijja air	Sperata aor (Hamilton, 1822)	VU	0.006	0.595
		Bujuri	Mystus tengara (Hamilton, 1822)	LC	2.816	1.142
	Schilbeidae	Bacha	Eutropiichthys vacha (Hamilton, 1822)	LC	0.045	0.152
		Batashi	Pachypterus atherinoides (Bloch, 1794)	LC	2.564	1.729
	Ailiidae	Kajuli	Ailia coila (Hamilton, 1822)	LC	0.079	0.607
	Siluridae	Boal	Wallago attu (Bloch and Schneider, 1801)	VU	0.071	13.098
		Boali pabda	Ompok bimaculatus (Bloch, 1794)	EN	0.059	0.303
		Pabda	Ompok pabo (Hamilton, 1822)	CR	0.379	1.078
		Modhu pabda	Ompok pabda (Hamilton, 1822)	EN	0.053	0.186
	Sisoridae	Baghair	Bagarius (Hamilton, 1822)	CR	0.015	0.653
	Chacidae	Chaka	Chaca (Hamilton, 1822)	EN	0.009	0.302
Mugiliformes	Mugilidae	Khorsula	Rhinomugil corsula (Hamilton, 1822)	LC	0.084	0.365
Cypriniformes	Botiidae	Rani	Botia dario (Hamilton, 1822)	EN	1.191	1.538
	Cobitidae	Gutum	Lepidocephalichthys guntea (Hamilton, 1822)	LC	1.641	1.426
	Cyprinidae	Dhela	Osteobrama cotio (Hamilton, 1822)	NT	0.832	0.459
		Sarpunti	Systomus sarana (Hamilton, 1822)	NT	0.056	1.254
		Mrigal	Cirrhinus cirrhosus (Bloch, 1795)	NT	0.001	0.088
		Common carp	Cyprinus carpio (Linnaeus, 1758)	-	0.021	1.214
		Tit punti	Pethia ticto (Hamilton, 1822)	VU	0.536	0.455
		Phutani punti	Pethia phutunio (Hamilton, 1822)	LC	0.398	0.251
		Jat punti	Puntius sophore (Hamilton, 1822)	LC	1.809	3.909
		Kalibaos/Kalia	Labeo calbasu (Hamilton, 1822)	LC	0.115	9.578
		Bata	Labeo bata (Hamilton, 1822)	NT	0.057	0.910
		Gonia	Labeo gonius (Hamilton, 1822)	NT	0.011	0.826
		Laccho	Cirrhinus reba (Hamilton, 1822)	NT	1.357	7.215
	Danionidae	Patharchata	Opsarius tileo (Hamilton, 1822)	EN	0.014	0.455
		Mola	Amblypharyngodon mola (Hamilton, 1822)	LC	1.669	1.465
		Ful chela	Salmostoma phulo (Hamilton, 1822)	NT	1.766	1.396
		Narkeli chela	Salmostoma bacaila (Hamilton, 1822)	LC	0.094	1.062
		Chhep chela	Devario (Hamilton, 1822)	VU	0.289	0.091
		Darkina	Esomus danrica (Hamilton, 1822)	LC	1.035	0.364
Cyprinidontiformes	Aplocheilidae	Kanpona	Aplocheilus Panchax (Hamilton, 1822)	LC	0.717	0.152
Anabantiformes	Nandidae	Meni	Nandus (Hamilton, 1822)	NT	0.006	0.067
	Anabantidae	Koi	Anabas testudineus (Bloch, 1792)	LC	0.119	0.228
Ovalentaria	Ambassidae	Gol chanda	Parambassis ranga (Hamilton, 1822)	LC	3.427	1.760
		Lomba chanda	Chanda nama (Hamilton, 1822)	LC	3.11	1.442
		Lal chanda	Parambassis lala (Hamilton, 1822)	LC	0.407	0.076
Beloniformes	Belonidae	Kankila	Xenentodon cancila (Hamilton, 1822)	LC	0.651	1.032
Gobiiformes	Gobiidae	Bele	Glossogobius giuris (Hamilton, 1822)	LC	0.303	0.419
Tetraodontiformes	Tetraodontidae	Potka	Leiodon cutcutia (Hamilton, 1822)	LC	0.159	3.186
Cichliformes	Cichlidae	Tilapia	Oreochromis niloticus (Linnaeus, 1758)	-	0.014	0.394
Osteoglossiformes	Notopteridae	Foli	Notopterus (Pallus, 1769)	VU	0.022	0.212
Clupeiformes	Clupeidae	Kachki	Corica soborna (Hamilton, 1822)	LC	2.405	0.365
		Chapila	Gudusia chapra (Hamilton, 1822)	VU	0.018	0.492
Synbranchiformes	Mastacembelidae	Guchi baim	Macrognathus pancalus (Hamilton, 1822)	LC	0.039	0.728
		Tara baim	Macrognathus aculeatus (Bloch, 1786)	NT	0.018	0.303
		Sal baim	Mastacembelus armatus (Lacepede, 1800)	EN	0.041	1.037
Decapoda	Palaemonidae	Gura chingri	Macrobrachium lamarrei (H. Milne Edwards, 1837)	LC	30.027	6.559
		Chatka icha	Macrobrachium malcolmsonii (H, Milne Edwards, 1844)	LC	37.179	6.958
		Golda	Macrobrachium rosenbergii (De Man, 1879)	LC	0.099	1.144
Total					100	100

NT = Near threatened, LC = Least concerned, VU = Vulnerable, EN = Endangered, DD = Data deficient, CR = Critically endangered.

3.3. Species diversity in different months

During the study period, variation was observed in species composition in different months. All the species were not continuously available throughout the study period. Monthly species richness was found to be the highest in December, followed by November (35), January (34), February (28), and March (25) (Figure 3).

Figure 3.

Monthly variation in species diversity.

3.4. Indices of diversity

The Shannon-Weiner diversity index (H′): The Shannon-Weiner diversity index (H′) was used to identify the biodiversity condition of the Shari-Goyain River in different months. The highest Shannon-Weiner diversity index indicates a high number of individuals and vice-versa. The species diversity index value was the highest in December (1.994), followed by January (1.959), February (1.864), March (1.827), and November (1.805). In November, species diversity was found to be the lowest (Figure 4).

Figure 4.

Values of different diversity indices in different months.

Margalef’s richness index (d): Margalef’s richness index counts the number of different species in a given area and is dependent on sampling size and effort. In the study area, Margalef’s richness value (d) was found to be the highest in December (3.430), and 3.120, 2.899, 2.552, and 2.325 in November, January, February, and March, respectively (Figure 4).

Pielou’s evenness index (J): Pielou’s evenness index reveals how individuals are distributed in a study area. Figure 4 shows the evenness index was 0.508, 0.544, 0.555, 0.554, and 0.561 in November, December, January, February, and March, respectively. The evenness index was the highest in March and the lowest in November.

Simpson’s dominance index (D): A maximum value was recorded in November (0.294) and a minimum was recorded in January (0.244) (Figure 4).

3.5. Production of fish from katha

A total of 76 individual catches were taken and the total production was 1,635.10 kg. The average estimated production of each katha catch was 21.51 ± 12.91 kg. It was found that on an averageeach katha was harvested 2.21 ± 0.78 times in that season. As total 187 kathas were harvested in the study area, the estimated total production was 8,889.98 kg.

The mean catch (kg) for each katha varied in different months. The individual average katha catch was the highest in November (41.09 ± 16.11 kg), followed by December (27.01 ± 12.21 kg), January (21.13 ± 9.60 kg), February (19.79 ± 11.58 kg), and March (12.42 ± 5.89 kg) (Figure 5). Variation in the mean individual katha catch was highly significant (p < 0.05) among the months (Table 2).

Figure 5.

Average catch (kg) per katha in different months.

Table 2.

One-way ANOVA table showing the variation of CPUA (kg/ha) and individual katha catch for 5 months (November 2018 to March 2019).

Mean scores	Source of variation	Sum of squares	df	Mean square	F	Sig. (p)
CPUA (kg/ha)	Between groups	167,473.639	4	41,868.410	14.796	0.000
	Within groups	200,907.994	71	2,829.690
	Total	368,381.633	75
Individual katha catch (kg)	Between groups	4,590.011	4	1,147.503	10.301	0.000
	Within groups	7,908.903	71	111.393
	Total	12,498.914	75

∗df = Degree of freedom, F = The F statistic used with ANOVA, Sig. = Significance.

The mean difference is significant at the 5% level.

3.6. Katha catch per unit effort

Analysis of katha catches was done for the months using catch monitoring records. The CPUE of katha is presented in Figure 6. The CPUE (kg/gear/ha/person/hour) was the highest in December (1.13 ± 0.37) followed by November (1.06 ± 0.4), January (0.8 ± 0.25), February (0.71 ± 0.23), and March (0.52 ± 0.21).

Figure 6.

CPUE (kg/gear/ha/person/hour) of katha

3.7. Katha catch per unit area (CPUA)

The CPUA (kg/ha) of katha is presented in Figure 7. CPUA was the highest in November (264.66 ± 18.21), followed by December (205.05 ± 27.77), January (175.02 ± 76.04), February (147.73 ± 52.11), and March (102.08 ± 41.04). The variation of CPUA among different months was found to be significantly different (p < 0.05) (Table 2).

Figure 7.

CPUA (kg/ha) of katha.

3.8. Different types of fishing gear are used in fishing in the Shari-Goyain River

Fishing gears used in the Shari-Goyain River were fishing nets, hooks and lines, wounding gear, and FADs. Different types of fish nets such as gill nets, seine nets, lift nets, cast nets, and dragged nets were used in this river. Hook and line, wounding gears such as koach, and traps such as katha were also used by the fishermen to catch fish.

3.9. Fishing nets and crafts used in katha fishing

Different types of fish nets and crafts were used in katha fishing of the Shari-Goyain River (Table 3). Three types of nets were used in katha fishing, such as the jhakijal (cast net), berjal (seine net), and hutarjal. The diameters of the jhakijal and hutarjal were 8.67 ± 0.63 m and 34.73 ± 26.18 m, respectively; the length and width of the berjal were 129.44 ± 11.86 m and 12.15 ± 1.08 m, respectively. There were small to medium-sized boats (dinghies) of 8.28 ± 1.21 m length and 1.40 ± 0.49 m width operated by one or two fishermen, and used during katha fishing. Dinghi was used to keep the fishes during harvesting and nets and to operate nets to catch fish.

Table 3.

Fishing nets and craft used in katha fishing.

	Local name	Length (m)	Width (m)	Diameter (m)
Net	Jhakijal	-	-	8.67 ± 0.63
	Berjal	129.44 ± 11.86	12.15 ± 1.08
	Hutarjal	-	-	34.73 ± 26.18
Craft	Dinghi	8.28 ± 1.21	1.40 ± 0.49	-

3.10. Impact of katha fishing in the Shari-Goyain River

A total of 56 respondents were interviewed who were engaged in fishing directly or indirectly. It was found that there were many positive and negative impacts of katha and katha fishing.

3.11. Positive impacts of katha

Total 56 fishers and non-fishers were interviewed about the positive impacts of katha fishing. All the respondents stated that fish used katha as a shelter, 57.14% of respondents stated that fish used katha as a nursing ground for juveniles, 83.93% of respondents said that kathas were good sources of food, and 92.86% of respondents thought that harvesting rates were higher in katha than that of openwater catch (Table 4).

Table 4.

Positive impacts of katha

Sl. no.	Impacts	Percentage (%)
1	Used as temporary shelter for fishes	100.00
2	Harvesting rate is higher than open water as the fishes are aggregated in katha	92.86
3	Good food source for fish	83.93
4	Used as nursing ground for juveniles	57.14

3.12. Negative impacts of katha fishing

Katha fishing has some negative impacts, which have been highlighted by the respondents (Table 5). Kruskal-Wallis One-way analysis of variance (ANOVA) by ranks was used to summarize respondents' perceptions of the major negative impacts of katha fishing. They were arranged in a numbering system ranging from 1 to 7, representing the highest to lowest ranked impacts on the Shari-Goyain River ecosystem, respectively. According to the perception of the fishers’, the most negative impact of katha fishing was reduction in open water fish catch (mean 92.50 units), followed by fish diversity decline (mean 140.50 units), damage of brood fishes due to indiscriminate fishing (mean 192.50 units), water pollution due to use of poison during katha fishing (mean 211.50 units), increasing siltation (mean 220.50 units), disturbing ecosystem (mean 246.50 units), and others (social conflict, boating problem, etc.) (Table 5). Notably, all the impacts identified and ranked by the respondents were statistically significant from each other.

Table 5.

Results of Kruskal-Wallis test for ranking the negative impacts of katha fishing.

Impacts	Mean	Rank
Reduction in open water fish catch	92.50	1
Fish diversity decline	140.50	2
Damage of brood fishes due to indiscriminate fishing	192.50	3
Water pollution due to use of poison during katha fishing	211.50	4
Increasing siltation	220.50	5
Disturbing ecosystem	246.50	6
Others (social conflict, boating problem, etc.)	271.50	7
Chi-square	101.95
Degree of freedom	6
Asymptotic significance	0.000

4. Discussion

4.1. The abundance of aquatic species in the katha

A total of 54 species of fish and prawn were harvested from the katha and Siluriformes (catfishes) was the most dominant order constituting about 39.123% of the total catch, followed by Cypriniformes (33.956%), and Decapoda (14.661%). Talukder et al. (2021) found 66 fish species in the Shari-Goyain River and catfishes (28.79%) dominated among the 14 groups. As the authors studied for one year and collected data from both katha and open water fishing, the number of available fish species was different to the present study. Ahmed and Akther (2008) and Uddin et al. (2015) also reported that Siluriformes was the dominant order in the brush shelter of the Titas River (47%) and Kaptai Lake (37%), respectively. The abundance of Siluriformes was due to food (prey) availability, a hazard-free environment, and shelter for breeding and nursing. Based on relative abundance (%), Macrobrachium malcolmsonii was the most abundant species in the capture (relative abundance of 37.179%), followed by M. lamarrei (30.027%) and Parambassis ranga (3.427%). Most of the catfishes are predatory and find available preys in the katha. Notably, Wallago attu topped the capture by bulk weight, accounting for 13.098% of the total katha catch in the Shari-Goyain River. Gut content analysis of this species harvested from katha was performed by Islam et al. (2006) and found 14 different prey items. Among the main diet categories, fish supplied 74.3% by weight and 80.9% by occurrence to the entire diet, prawns 18.5% by weight and 11.0% by occurrence to the total diet, and plants 7.2% by weight and 8.1% by occurrence to the total diet (Islam et al., 2006). Despite the great degree of seasonality in katha fisheries, they are often controlled by a number of predators due to the abundance of their food in and around the kathas. The trophic diversity of kathas provides ideal feeding conditions for herbivores and planktivores, attracting high numbers of carnivores and piscivores. As a result, a specific food web based on kathas emerges, starting with herbivory and ending with piscivory. Wallago attu is an aggressive and voracious predator that could have an impact on the prey fish species' community structure and population growth (Islam et al., 2006). On the other hand, Rahman et al. (2016a) and Miah (2012) observed that Cypriniformes was dominant in Hakaluki Haor contributed 36% and 54%, respectively. Uddin et al. (2015) also found one exotic species (tilapia) and 2 species of prawn from the katha during the harvesting period. All these findings coincided with the present study.

4.2. Production from katha fishing

Fishing was done by hand picking, using wounding gear such as koach (spear), nets such as seine nets and cast nets with an average mesh size of 1.04 ± 0.85 cm. This mesh size is very small, so that undersized fish are also captured, and these kinds of nets are illegal for harvesting. The average katha catch was highest in November and lowest in March and the variation in the mean individual katha catch was significantly different (p < 0.05) among the months. In November, the average catch was higher because at that time, katha fishing had just started as the water started receding and fish availability was higher. The katha catch decreased as the intensity of katha fishing increased, reaching its lowest point in March. It might be due to the gradual decrease of total stock in the river. Notably, the monsoon (June–September) is the breeding season for almost all fish species in Bangladesh, and no natural recruitment occurs during the katha fishing period. Again, there was no chance to enter any fish from the nearby waterbodies as those were dried up or detached from the river, and there was less chance to migrate fish from the Surma River as it is deeper than the Shari-Goyain River. CPUA was the highest in November because water depth was lowering and katha harvesting started when fish stock was available in the river. The declination in CPUA (kg/ha) from November to the following months may be the resulted of increasing fishing pressure from katha fishing and open water catch. The CPUE (kg/gear/ha/person/hour) was found to be the highest in December (1.13 ± 0.37) and the lowest in March (0.52 ± 0.21). This variation in CPUE occurred due to fishing pressure in the earlier months, and as the water level was very low, fish migration stopped. To the best of the authors’ knowledge, no references dealing with the CPUE of katha are available to validate and compare with this study.

4.3. Monthly variation of diversity indices

The Shannon-Wiener diversity index considers the richness and proportion of each species. According to the Shannon-Weiner diversity index (H′), the species diversity was the highest in December and the lowest in November. As in the case of the Shannon-Weiner diversity index (H′), the highest diversity index indicates high individuality and the lowest diversity indicates a low number of individuals. Diversity index was lowest in November because small numbers of katha were harvested in November. Species diversity decreased after December because the water level started receding and the harvesting rate was also higher in December. Mustafa (2017) recorded Shannon-Weiner diversity index (H′) values of 2.77, 2.98, 2.92, 2.89, 2.81, and 2.85 in the Titas River from 1997 to 2002. Iqbal et al. (2015) recorded Shannon-Weiner diversity index (H′) values ranging from 1.8 to 3.40 in the Hakaluki Haor and Hossain et al. (2017) found H′ values ranging from 2.07 to 2.41 in the Kushiara River. Chowdhury et al. (2019) recorded that the Shannon-Weiner diversity index fluctuated between 2 and 2.5, with a mean value of 2.30 ± 0.14 in the Surma River. In the present study, H′ ranged from 1.805 to 1.994, which indicates the less diversified fish population in the Shari-Goyain River. It might be related to changes in geographical location, the effects of coal mining drainage, fish biological condition, water quality, different fishing techniques, and harvesting frequency. The Margalef’s richness index was the highest in December (3.430) and the lowest in March (2.325). That means the sample size of fish was the highest in December and also the number of individuals, as new individuals were added to the stocks, which increased the species richness. Siddique et al. (2016) recorded the average highest richness (d) value of 8.39 in December and the lowest, 4.53, in July in the entire Chalan Beel. Galib et al. (2013) recorded a richness value that varied from 6.973 in June to 8.932 in November for the Choto Jamuna River in Bangladesh. According to Nair et al. (1989), to some extent, the Margalef’s richness index may be different from the actual diversity value because it depends on sample size, not on the evenness and species richness of the data.

The evenness index was calculated at 0.508, 0.544, 0.555, 0.554, and 0.561 in November, December, January, February, and March, respectively. The study revealed that the distribution of species in the Shari-Goyain River was equally distributed in different months. Hossain et al. (2017) found Pielou’s evenness index (J) between 0.99 and 1.15 in the Kushiara River and Iqbal et al. (2015) recorded evenness index fluctuating from 0.79 to 0.9 for the Surma River. Jannatul et al. (2015) showed an evenness index of 0.36–0.76 for the Halda River. Chowdhury et al. (2019) found the maximum Pielou’s evenness index value was 2.2 in April and the minimum value was 1.47 in July, with a mean value of 1.93 ± 0.23 for the Surma River. Dominance index represents the fraction of common species. Simpson’s dominance index (d) values were the highest in November (0.294) and the lowest in January (0.244) found in the Shari-Goyain River. The greater the index value, the greater the sample diversity of the dominant species suppressing others. Hossain et al. (2017) recorded the Simpson’s dominance index at 2.78 to 7.23 for the Kushiara River. Chowdhury et al. (2019) recorded the highest value of the Simpson’s dominance index at 7.98 in August and the lowest value of 5.32 in October, with an average value of 6.99 ± 0.86. Rahman et al. (2016b) recorded a dominance index of 0.064 to 0.0133 in the Bishkhali River of Barguna district. The possible causes of the difference in the diversity indices during the study period might be the high intensity of fishing pressure and the drastic reduction of water from the river.

4.4. Impact of katha fishing in the Shari-Goyain River

During the study period, 38 species were found available in December, followed by November (35), January (34), February (28), and March (25). Significant differences in species availability were noticed due to gradual lowering of water depth, high harvesting frequency, efficiency of gears, deterioration of water quality, and the biological condition of fishes. Islam et al. (2019) observed 49 and 39 fish species in November and December, respectively in the Juri River of Sylhet district, which are higher than those of the present study. Possible reasons behind this difference are variations in location and the catch was from open river catch, not only from katha. In the study area, katha was harvested 2.21 ± 0.78 times after establishment in a season. Mustafa (2017) also found that a single katha was harvested 2 to 3 times repeatedly in a season, which supports the present study. Various katha materials were used in katha, such as bamboo, branches of trees, etc. which make it a temporary shelter where fish can take shelter, and be used as a feeding, breeding, and nursing ground (Uddin et al., 2015; Sharma et al., 2014; Joadder et al., 2016; Kunda et al., 2022). Katha materials acted as substrates for periphyton growth, which is used as fish food. Thus, katha provides a short-term shelter and food for various types of fishes (Mustafa, 2017). For the above reasons, many threatened species were found available in the katha. On the classification table, all species are classified according to the IUCN Red List in Bangladesh (IUCN Bangladesh, 2015), and a total of 16 threatened fish species are found, including 2 critically endangered, 7 endangered, and 7 vulnerable. This is an encouraging sign that the katha shelter can be helpful for the conservation of threatened fish species if it is not harvested.

On the other hand, the main perceived negative consequence of katha fishing is a decrease in open water capture fisheries, followed by a decline in fish diversity, destruction of brood fishes, water pollution owing to poison use during katha fishing, increased siltation, and disturbance to the ecosystem. Uddin et al. (2015) also found katha fishing destructive because of indiscriminate and complete harvest of katha. Mustafa (2017) also stated that katha fishing is a destructive fishing method. Sultana et al. (2017) found katha fishing as a threat to Bhawal Beel biodiversity. Galib et al. (2009), Pandit et al. (2015), Rahman et al. (2016a) and Pandit et al. (2021) also stated the negative impact of katha fishing, which coincided with the present study. Though this type of fishing is prohibited according to fisheries regulations in Bangladesh (Joadder et al., 2016), its intensity in the Shari-Goyain River is increasing day by day. According to the Protection and Conservation of Fish Act (1950), the construction of permanent or temporary structures in the wetlands, the use of nets of below prescribed mesh sizes, in addition to the catching of undersized fish, are strictly prohibited. However, the poor people engaged in katha fishing are compelled to disregard the regulations to meet their financial and basic needs, which are not monitored by the respective authorities. As of now, this is more of a preliminary description of katha fishing and the biodiversity associated with a fished katha. However, it provides the message that to conserve species diversity in the river, katha fishing should be stopped or converted into permanent fish sanctuaries (Kunda et al., 2022).

5. Conclusion

The Shari-Goyain River is an important river in Bangladesh, and katha fishing is very common in this river, which has impacts on fish biodiversity. Soon after the monsoon, the water level in this river began to fall, and from January to March, there was insufficient water to provide shelter, feeding, and nursing for fish, except in a few deeper sections of the river. Kathas are being operated in the deeper portions, and fish enter the katha for their shelter and are badly harvested by the fishers, resulting in very few of the brood fish being left for next year’s breeding. Thus, this fishing method increases the vulnerability of rare and endangered fish species in the river. According to the key informant and the observation of our study team, katha fishing should be stopped and several effective sanctuaries need to be established to revive and enhance fish biodiversity in the Shari-Goyain River.

Declarations

Author contribution statement

Sumi Rani Das: Performed the experiments; Analyzed and interpreted the data; Wrote the paper.

Debasish Pandit: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Wrote the paper.

Ahmed Harun-Al-Rashid: Analyzed and interpreted the data; Wrote the paper.

Nishat Tasnim: Performed the experiments.

Mrityunjoy Kunda: Conceived and designed the experiments; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Funding statement

Professor Dr. Mrityunjoy Kunda was supported by the Bangladesh Agricultural Research Council (BARC), Dhaka, Bangladesh through National Agricultural Technology Program: Phase II Project (Sub-Project ID #035).

Data availability statement

Data will be made available on request.

Declaration of interest’s statement

The authors declare no conflict of interest.

Additional information

No additional information is available for this paper.